Date: Sat, 18 Apr 2026 04:19:01 -0400
The AI Integration Strategy Meeting discussed the challenges and strategies
for integrating AI into C++. Key points included the need for a unified
direction for AI in C++, emphasizing the importance of statistics and graph
libraries. The meeting highlighted the role of AI in increasing
productivity and the need for careful review and understanding of
AI-generated code. The discussion also covered the importance of succession
planning, with a focus on identifying and engaging new leaders to continue
the work. Additionally, the meeting addressed the potential use of AI in
committee papers and the need for clear guidelines to avoid issues of
academic integrity. The AI Integration Strategy Meeting focused on
re-engaging with Andres, a valued but currently disengaged committee
member. Speaker 1 suggested a three-pronged approach: internal delegation,
public acknowledgment of Andres' upcoming retirement, and recruiting new
lead co-authors. Speaker 2 confirmed a two-year retirement window and
agreed to pre-communicate this within the group. They discussed the
potential benefits and drawbacks of using the SG 19 instant messaging
channel, ultimately deciding not to create it due to the additional
workload. The meeting concluded with a positive note on the discussed
topics and future plans.Action Items
- MW Draft the AI directions addition for the Directions Group (covering
agentic coding guidance, AI use in committee work, AI-cheating guidance,
and C++ as AI infrastructure) and share slides/draft with the group for
input (target: next week / over next couple weeks)
- PR Reach out to Andre (Andre Kaminsky) to re-engage him on the graph
work and SG19 review, and follow up on his earlier comments to bring him
back into the effort
- PR Continue implementing graph v3 (restart graph implementation), add
unit tests, and progress the descriptors approach so it can be presented to
SG19 and LWG as needed
- PR Talk with your director about using the proposed C++ middle-layer
approach under PyTorch (evaluate integration and company plans)
- PR Discuss with your manager about identifying company colleagues who
can take on or co-author the graph/library work to ensure continuity and
increased involvement
Outline.Graph Library and AI Integration Challenges
- PR shares his experience with implementing graph v3 and the challenges
of getting Andrew's participation.
- MW advises Speaker 2 to present both versions of the paper to LEWG to
ensure a balanced approach.
- PR discusses the use of agentic coding for graph data structures and
the benefits of using AI for test case generation.
- MW emphasizes the importance of using AI cautiously and reviewing
every line of code generated by AI.
AI in C++ and Future Directions
- MW discusses the importance of AI in C++ and the need for a unified
direction for AI integration.
- MW outlines the three layers of the C++ AI stack: data science
foundation, core structure, and execution parallelism.
- MW emphasizes the need for a data frame to translate between C++ and
Python data structures.
- MW discusses the importance of having a multi-backend architecture for
AI in C++ to support various hardware platforms.
Succession Planning and Future Commitment
- MW suggests involving more co-authors and promoting others to take
over the leadership role.
- PR mentions the need for a Beman project or Boost implementation to
gain more usage experience for the graph library.
- PR advises Speaker 2 to involve his company's delegates and other
members of the network group to ensure the continuity of the project.
Re-Engaging with Andres and Addressing Feedback
- MW discusses the importance of re-engaging with Andres, noting his
valuable feedback and non-aggressive approach.
- MW suggests apologizing to Andres for being unresponsive due to busy
work schedules and emphasizes the need to maintain good relationships.
On Wed, Jan 14, 2026 at 11:43 AM Michael Wong <fraggamuffin_at_[hidden]>
wrote:
> Hi all,
>
> there will be an SG19 meeting 2 PM ET this week to replace normally the
> one last week as I assumed people might still be on holiday, focused on
> Graphs and Stats, We also like to use this kick off year beginning look
> at general machine learning proposals for the future. Thank you.
>
> Phil Ratzloff is inviting you to a scheduled Zoom meeting.
>
> Topic: SG19 Machine Learning
> Time: Jan 15, 2026 02:00 PM Eastern Time (US and Canada)
>
> Join from PC, Mac, Linux, iOS or Android:
> https://iso.zoom.us/j/94859326954?pwd=CieTHKjbEDGszHCkSpDdqnNQtJQDTK.1
> Password: 994533
>
> Or iPhone one-tap :
> US: +17866351003,,94859326954# or +12678310333,,94859326954#
> Or Telephone:
> Dial(for higher quality, dial a number based on your current location):
> US: +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1
> 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805
> or +1 646 876 9923 or +1 651 372 8299 or +1 669 219 2599 or +1 669 900
> 6833 or +1 720 928 9299 or +1 971 247 1195 or +1 206 337 9723 or +1 213
> 338 8477 or +1 253 215 8782 or +1 346 248 7799 or +1 408 638 0968 or +1
> 602 753 0140 or 877 853 5247 (Toll Free) or 888 788 0099 (Toll Free)
> Meeting ID: 948 5932 6954
> Password: 994533
> International numbers available: https://iso.zoom.us/u/aAD06MvSf
>
>
> Agenda:
>
> 1. Opening and introductions
>
> The ISO Code of conduct:
> https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100397.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100397.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzoxY2FlOjUzNmY1NzIzY2Q5N2Y0ZTM0Y2FhMjk3NTFjYTAyNzhiMmEwYzYwMjM0ZDViMGRjNGZiOTgzNmE0YmE0Y2ZlYWI6aDpUOk4>
>
> IEC Code of Conduct:
>
> https://www.iec.ch/basecamp/iec-code-conduct-technical-work
> <https://protect.checkpoint.com/v2/r01/___https://www.iec.ch/basecamp/iec-code-conduct-technical-work___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzpjYWE3OjlkMzFlMDg3OTYwOTdkZDQ2Yjg1NzBlZTFmMTg2NTc5ZjNiZDAwZTg4YzU0OWJjMDY4ZTcwZGYyMmUwNzg5Mjk6aDpUOk4>
>
> ISO patent policy.
>
>
> https://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/Common_Policy.htm?nodeid=6344764&vernum=-2
> <https://protect.checkpoint.com/v2/r01/___https://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/Common_Policy.htm?nodeid=6344764&vernum=-2___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzowNmFhOjc2NzUxNDU3YzliYmFhNTczZDQ3ZGU3ODJkZDAyNjI0MWFlZGU1MjQzMGZhNjljYTI3OWE4ZDVlZWNiNWEyYTA6aDpUOk4>
>
> The WG21 Practices and Procedures and Code of Conduct:
>
> https://isocpp.org/std/standing-documents/sd-4-wg21-practices-and-procedures
> <https://protect.checkpoint.com/v2/r01/___https://isocpp.org/std/standing-documents/sd-4-wg21-practices-and-procedures___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6Nzo2MjE0OmU4NGQ3NjQyMTA2ZDU2NGM1M2ZmZTdiODdmNjNhN2E1Y2JmNmFlY2VlM2Q2ZjZhYzRiNjI2ZTI0MTk2OTMxODg6aDpUOk4>
>
> 1.1 Roll call of participants
>
> 1.2 Adopt agenda
>
> 1.3 Approve minutes from previous meeting, and approve publishing
> previously approved minutes to ISOCPP.org
>
> 1.4 Action items from previous meetings
>
> 2. Main issues (125 min)
>
> 2.1 General logistics
>
> 2.2 Paper reviews
>
> 2.2.1: ML topics
>
> 2.2.1.1 Statistics Proposals
> P1708 Basic Statistics
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p1708r9.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p1708r9.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzpmZGRjOmZkM2M3ZTQ0ZDY1MDgwZGVhMGNkNzllZGQ0Yjk4MDExNDg3MjI2OWJjMjU5N2I4MjVmN2U5OTBkYTJlYTYwZGM6aDpUOk4>
> P2681 More Basic Statistics
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/p2681r1.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/p2681r1.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzowZDUyOmY0ZjU5NGY5Y2UwZGUwYzlkYzQxNDA3ZTk1Njk0OTk1YmM4YmM5OTEyMjVhNWM0NmZmY2RlOTNiNjc0N2Q3ZTU6aDpUOk4>
>
>
> 2.2.1.1 Graph Proposal Review
>
> P3126 Graph Library Overview https://wg21.link/P3126
> P3127 Graph Library Background and Terminology https://wg21.link/P3127
> P3128 Graph Library Algorithms https://wg21.link/P3128
> P3129 Graph Library Views https://wg21.link/P3129
> <https://wg21.link/P31298>
> P3130 Graph Library Graph Container Interface https://wg21.link/P3130
> <https://wg21.link/P31298>
> P3131 Graph Library Graph Containers https://wg21.link/P3131
> <https://wg21.link/P31298>
> P3337 Graph Comparison to other graph libraries https://wg21.link/P3337
> <https://wg21.link/P31298>
>
> 3. Any other business
>
> 4. Review
>
> 4.1 Review and approve resolutions and issues [e.g., changes to SG's
> working draft]
>
> 4.2 Review action items (5 min)
>
> 5. Closing process
>
> 5.1 Establish next agenda
>
>
> 5.2 Future meeting
> Jan 15, 2026 02:00 PM ET: Statistics
> Feb 12, 2026 02:00 PM ET: Graph
> Mar 12, 2026 02:00 PM ET: Statistics; F2F London
> Apr 9, 2026 02:00 PM ET: Graph
> May 14, 2026 02:00 PM ET: Statistics
> June 11, 2026 02:00 PM ET: F2F cancelled
> July 9, 2026 02:00 PM ET: Graph
> Aug 13, 2026 06:00 PM ET: Statistics
> Sep 10, 2026 02:00 PM ET: CPPCON 9/12-18
> Oct 8, 2026 02:00 PM ET: Graph
> Nov 12, 2026 02:00 PM ET: F2F, cancelled
> Dec 10, 2026 02:00 PM ET: Statistics
>
>
for integrating AI into C++. Key points included the need for a unified
direction for AI in C++, emphasizing the importance of statistics and graph
libraries. The meeting highlighted the role of AI in increasing
productivity and the need for careful review and understanding of
AI-generated code. The discussion also covered the importance of succession
planning, with a focus on identifying and engaging new leaders to continue
the work. Additionally, the meeting addressed the potential use of AI in
committee papers and the need for clear guidelines to avoid issues of
academic integrity. The AI Integration Strategy Meeting focused on
re-engaging with Andres, a valued but currently disengaged committee
member. Speaker 1 suggested a three-pronged approach: internal delegation,
public acknowledgment of Andres' upcoming retirement, and recruiting new
lead co-authors. Speaker 2 confirmed a two-year retirement window and
agreed to pre-communicate this within the group. They discussed the
potential benefits and drawbacks of using the SG 19 instant messaging
channel, ultimately deciding not to create it due to the additional
workload. The meeting concluded with a positive note on the discussed
topics and future plans.Action Items
- MW Draft the AI directions addition for the Directions Group (covering
agentic coding guidance, AI use in committee work, AI-cheating guidance,
and C++ as AI infrastructure) and share slides/draft with the group for
input (target: next week / over next couple weeks)
- PR Reach out to Andre (Andre Kaminsky) to re-engage him on the graph
work and SG19 review, and follow up on his earlier comments to bring him
back into the effort
- PR Continue implementing graph v3 (restart graph implementation), add
unit tests, and progress the descriptors approach so it can be presented to
SG19 and LWG as needed
- PR Talk with your director about using the proposed C++ middle-layer
approach under PyTorch (evaluate integration and company plans)
- PR Discuss with your manager about identifying company colleagues who
can take on or co-author the graph/library work to ensure continuity and
increased involvement
Outline.Graph Library and AI Integration Challenges
- PR shares his experience with implementing graph v3 and the challenges
of getting Andrew's participation.
- MW advises Speaker 2 to present both versions of the paper to LEWG to
ensure a balanced approach.
- PR discusses the use of agentic coding for graph data structures and
the benefits of using AI for test case generation.
- MW emphasizes the importance of using AI cautiously and reviewing
every line of code generated by AI.
AI in C++ and Future Directions
- MW discusses the importance of AI in C++ and the need for a unified
direction for AI integration.
- MW outlines the three layers of the C++ AI stack: data science
foundation, core structure, and execution parallelism.
- MW emphasizes the need for a data frame to translate between C++ and
Python data structures.
- MW discusses the importance of having a multi-backend architecture for
AI in C++ to support various hardware platforms.
Succession Planning and Future Commitment
- MW suggests involving more co-authors and promoting others to take
over the leadership role.
- PR mentions the need for a Beman project or Boost implementation to
gain more usage experience for the graph library.
- PR advises Speaker 2 to involve his company's delegates and other
members of the network group to ensure the continuity of the project.
Re-Engaging with Andres and Addressing Feedback
- MW discusses the importance of re-engaging with Andres, noting his
valuable feedback and non-aggressive approach.
- MW suggests apologizing to Andres for being unresponsive due to busy
work schedules and emphasizes the need to maintain good relationships.
On Wed, Jan 14, 2026 at 11:43 AM Michael Wong <fraggamuffin_at_[hidden]>
wrote:
> Hi all,
>
> there will be an SG19 meeting 2 PM ET this week to replace normally the
> one last week as I assumed people might still be on holiday, focused on
> Graphs and Stats, We also like to use this kick off year beginning look
> at general machine learning proposals for the future. Thank you.
>
> Phil Ratzloff is inviting you to a scheduled Zoom meeting.
>
> Topic: SG19 Machine Learning
> Time: Jan 15, 2026 02:00 PM Eastern Time (US and Canada)
>
> Join from PC, Mac, Linux, iOS or Android:
> https://iso.zoom.us/j/94859326954?pwd=CieTHKjbEDGszHCkSpDdqnNQtJQDTK.1
> Password: 994533
>
> Or iPhone one-tap :
> US: +17866351003,,94859326954# or +12678310333,,94859326954#
> Or Telephone:
> Dial(for higher quality, dial a number based on your current location):
> US: +1 786 635 1003 or +1 267 831 0333 or +1 301 715 8592 or +1
> 312 626 6799 or +1 470 250 9358 or +1 470 381 2552 or +1 646 518 9805
> or +1 646 876 9923 or +1 651 372 8299 or +1 669 219 2599 or +1 669 900
> 6833 or +1 720 928 9299 or +1 971 247 1195 or +1 206 337 9723 or +1 213
> 338 8477 or +1 253 215 8782 or +1 346 248 7799 or +1 408 638 0968 or +1
> 602 753 0140 or 877 853 5247 (Toll Free) or 888 788 0099 (Toll Free)
> Meeting ID: 948 5932 6954
> Password: 994533
> International numbers available: https://iso.zoom.us/u/aAD06MvSf
>
>
> Agenda:
>
> 1. Opening and introductions
>
> The ISO Code of conduct:
> https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100397.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.iso.org/files/live/sites/isoorg/files/store/en/PUB100397.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzoxY2FlOjUzNmY1NzIzY2Q5N2Y0ZTM0Y2FhMjk3NTFjYTAyNzhiMmEwYzYwMjM0ZDViMGRjNGZiOTgzNmE0YmE0Y2ZlYWI6aDpUOk4>
>
> IEC Code of Conduct:
>
> https://www.iec.ch/basecamp/iec-code-conduct-technical-work
> <https://protect.checkpoint.com/v2/r01/___https://www.iec.ch/basecamp/iec-code-conduct-technical-work___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzpjYWE3OjlkMzFlMDg3OTYwOTdkZDQ2Yjg1NzBlZTFmMTg2NTc5ZjNiZDAwZTg4YzU0OWJjMDY4ZTcwZGYyMmUwNzg5Mjk6aDpUOk4>
>
> ISO patent policy.
>
>
> https://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/Common_Policy.htm?nodeid=6344764&vernum=-2
> <https://protect.checkpoint.com/v2/r01/___https://isotc.iso.org/livelink/livelink/fetch/2000/2122/3770791/Common_Policy.htm?nodeid=6344764&vernum=-2___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzowNmFhOjc2NzUxNDU3YzliYmFhNTczZDQ3ZGU3ODJkZDAyNjI0MWFlZGU1MjQzMGZhNjljYTI3OWE4ZDVlZWNiNWEyYTA6aDpUOk4>
>
> The WG21 Practices and Procedures and Code of Conduct:
>
> https://isocpp.org/std/standing-documents/sd-4-wg21-practices-and-procedures
> <https://protect.checkpoint.com/v2/r01/___https://isocpp.org/std/standing-documents/sd-4-wg21-practices-and-procedures___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6Nzo2MjE0OmU4NGQ3NjQyMTA2ZDU2NGM1M2ZmZTdiODdmNjNhN2E1Y2JmNmFlY2VlM2Q2ZjZhYzRiNjI2ZTI0MTk2OTMxODg6aDpUOk4>
>
> 1.1 Roll call of participants
>
> 1.2 Adopt agenda
>
> 1.3 Approve minutes from previous meeting, and approve publishing
> previously approved minutes to ISOCPP.org
>
> 1.4 Action items from previous meetings
>
> 2. Main issues (125 min)
>
> 2.1 General logistics
>
> 2.2 Paper reviews
>
> 2.2.1: ML topics
>
> 2.2.1.1 Statistics Proposals
> P1708 Basic Statistics
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p1708r9.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2024/p1708r9.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzpmZGRjOmZkM2M3ZTQ0ZDY1MDgwZGVhMGNkNzllZGQ0Yjk4MDExNDg3MjI2OWJjMjU5N2I4MjVmN2U5OTBkYTJlYTYwZGM6aDpUOk4>
> P2681 More Basic Statistics
> https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/p2681r1.pdf
> <https://protect.checkpoint.com/v2/r01/___https://www.open-std.org/jtc1/sc22/wg21/docs/papers/2023/p2681r1.pdf___.YzJ1OnNhc2luc3RpdHV0ZTpjOm86MGIxYjQ3Yzg2NDg1MzZkMGU1M2VhOGEwNWM3YTc3YWE6NzowZDUyOmY0ZjU5NGY5Y2UwZGUwYzlkYzQxNDA3ZTk1Njk0OTk1YmM4YmM5OTEyMjVhNWM0NmZmY2RlOTNiNjc0N2Q3ZTU6aDpUOk4>
>
>
> 2.2.1.1 Graph Proposal Review
>
> P3126 Graph Library Overview https://wg21.link/P3126
> P3127 Graph Library Background and Terminology https://wg21.link/P3127
> P3128 Graph Library Algorithms https://wg21.link/P3128
> P3129 Graph Library Views https://wg21.link/P3129
> <https://wg21.link/P31298>
> P3130 Graph Library Graph Container Interface https://wg21.link/P3130
> <https://wg21.link/P31298>
> P3131 Graph Library Graph Containers https://wg21.link/P3131
> <https://wg21.link/P31298>
> P3337 Graph Comparison to other graph libraries https://wg21.link/P3337
> <https://wg21.link/P31298>
>
> 3. Any other business
>
> 4. Review
>
> 4.1 Review and approve resolutions and issues [e.g., changes to SG's
> working draft]
>
> 4.2 Review action items (5 min)
>
> 5. Closing process
>
> 5.1 Establish next agenda
>
>
> 5.2 Future meeting
> Jan 15, 2026 02:00 PM ET: Statistics
> Feb 12, 2026 02:00 PM ET: Graph
> Mar 12, 2026 02:00 PM ET: Statistics; F2F London
> Apr 9, 2026 02:00 PM ET: Graph
> May 14, 2026 02:00 PM ET: Statistics
> June 11, 2026 02:00 PM ET: F2F cancelled
> July 9, 2026 02:00 PM ET: Graph
> Aug 13, 2026 06:00 PM ET: Statistics
> Sep 10, 2026 02:00 PM ET: CPPCON 9/12-18
> Oct 8, 2026 02:00 PM ET: Graph
> Nov 12, 2026 02:00 PM ET: F2F, cancelled
> Dec 10, 2026 02:00 PM ET: Statistics
>
>
Received on 2026-04-18 08:19:16
